Nair et al.
850 Journal of Medicinal Chemistry, 1979, Vol. 22, No. 7
Folate Analogues Altered in the C9-N10Bridge Region: 11-Thiohomofolic Acid M. G. Nair,* Shiang-Yuan Chen, Department of Biochemistry, College of Medicine, University o f South Alabama, Mobile, Alabama 36688
Roy L. Kisliuk, Y. Gaumont, and D. Strumpf Departments of Biochemistry and Pharmacology, School of Medicine, Tufts University, Boston, Massachusetts 201 11. Received December 27, 1978
The synthesis of 11-thiohomofolicacid (2) has been accomplished by an unambiguous procedure. Reaction of l-chloro-4-[p-(carbomethoxy)thiophenoxy]-2-butanone (10) with hydroxylamine under carefully controlled conditions gave the corresponding oxime 33. Conversion of this oxime to l-phthalimido-4-[p-(carbomethoxy)thiophenoxy]-2-butanoneoxime (4) was carried out by its reaction with potassium phthalimide using crown 18 ether as a catalyst. Hydrazinolysis of compound 4 gave l-amino-4-[p-(carbomethoxy)thiophenoxy]-2-butanone oxime ( 5 ) , which was used for the construction of the title compound 2 by modification of the Boon and Leigh procedure. An alternate synthesis utilizing l-hydroxy-4[p-(carbmethoxy)thiophenoxy] -2-butanone (11) and 4-hydroxy-2,5,6-triainopyrimidine has also been carried out. Compound 2 did not exhibit any antifolate activity against Lactobacillus casei or Streptococcus faecium. The dithionite reduction product, 7,8-dihydro-ll-thiohomofolic acid, was able to function as a substrate of L. casei dihydrofolate reductase. The catalytic reduction product of 2, consisting of a mixture of diastereomers, exhibited powerful antifolate activity against both these organisms. Several years ago, DeGraw and co-workers' synthesized an analogue of folic acid in which an extra methylene group ' position of the was inserted between the C9 and the NO vitamin. This compound, homofolic acid (l),showed some very interesting biological properties and was the subject of extensive investigation during the past decade. For example, 7,8-dihydrohomofolic acid was shown to be a substrate of L. casei dihydrofolate reductase, and its tetrahydro form was shown to be a powerful inhibitor of thymidylate synthetase.2 Dihydrohomofolate substituted for folate as a growth factor for S. faecium, while the tetrahydro derivative antagonized its growth. This inhibition of growth was later attributed to the d , ~ diastereomer3 of tetrahydrohomofolic acid. The enzymatic reduction product of 7,8-dihydrohomofolic acid, i.e., 1 , ~ tetrahydrohomofolic acid, was capable of acting as a pseudocofactor of thymidylate synthesis in S. faecium and L. casei.4 Mishra5 and co-workers have shown that dihydrohomofolate was active against an antifolate-resistant leukemia L1210/FR8, which contains high levels of dihydrofolate reductase, and that the tetrahydro derivative prolonged the life span of MTX-resistant leukemic miceS6 The lack of antifolate activity of 7,8-dihydro and 1 , ~ tetrahydro derivatives of homofolic acid in certain microorganisms could be explained in terms of their ability to be metabolized to N5,N'1-methylenetetrahydrohomofolate and its subsequent participation as a pseudocofactor in thymidylate b i o ~ y n t h e s i s . ~Therefore, ,~ we were interested in designing a potential substrate of dihydrofolate reductase which, in its enzymatically reduced tetrahydro form, is unable to carry cyclic 1 carbon fragments in the bridge region, This molecular modification would effectively prevent the molecule from participating as a pseudocofactor in thymidylate synthesis. Therefore, the enzymatically reduced tetrahydro derivative of this homofolate analogue was envisioned to be potentially capable of interfering with tetrahydrofolate utilization. Substitution of a sulfur atom for the 11-amino group of homofolic acid appeared to meet these requirements. As part of a continuing program aimed at developing analogues of folic acid which are alterd in the C9-N10 bridge regi~n,~-lO this paper details the synthesis and preliminary biological evaluation of 11-thiohomofolic acid. Chemistry. The partial side chain 5 could be obtained from 4 by hydrazinoly~is.~,~ Therefore, initial attempts were directed toward the synthesis of 3 and 4. The preparation of p-(carbomethoxy)thiophenol from p -
aminobenzoic acid was accomplished using earlier procedures.l1,l2 We proposed to prepare 3 by the reaction of potassium phthalimide with chloromethyl ketone 10, which in turn could be prepared by the Michael addition of p-(carbomethoxy)thiophenol and hydroxymethyl vinyl ketone (15) and the subsequent treatment of the product with thionyl chloride. To optimize conditions for this reaction, the Michael addition of this thiophenol with acrolein and methyl vinyl ketone was examined as models which resulted in the formation of compounds 19 and 21. Conversions of 1,4-butynediol (14) to 15, 16, and 17 were carried out using HgS04 as a catalyst and using the appropriate reaction ~0nditions.l~Michael addition of p (carbomethoxy)thiophenol to these substituted vinyl ketones gave the hydroxymethyl, chloromethyl, and acetoxymethyl ketones 11, 10, and 12, respectively. The chloromethyl ketone 10 can also be prepared from 20 by converting it successively to the acid chloride, diazo ketone, and chloromethyl ketone by the Amdt-Eistert pr0~edure.l~ Thus, treatment of p-(carbomethoxy)thiophenol with P-propiolactone or iodopropionic acid resulted in the formation of 20, which was converted to the acid chloride 23b by treatment with thionyl chloride. This acid chloride was treated with diazomethane, and the diazo ketone thus formed was reacted with gaseous HC1 to produce 10 in good yield. Compounds 9, 11, and 12 were also prepared as potential starting materials for 10, which in turn could be converted to 3 by reaction with potassium phthalimide. A more direct approach to the synthesis of 5 or 6 consisted of converting 10 to the corresponding keto azide 13, protection of the carbonyl group of 13 as the oxime or ketal, and subsequent reduction of the azide to a primary amine. Although these reactions appeared straightforward, treatment of 10 with sodium azide in aqueous acetone resulted in the formation of a major product whose structure was established as 24 by mass spe~trometry'~ and NMR. Therefore, attempts were made to prepare 3 from 10 by reaction with potassium phthalimide using acetonitrile as a solvent and crown 18 ether as a catalyst. The molecular ion of this reaction product had an m l e value of 38324but had the rearranged structure 25 rather than the expected 3.15 Since the occurrence of this rearrangementI5 was discovered only after experiencing the consistent failures to ring close the dithionite reduction product of either compound 31 or 32 to a pteridine, all the pyrimidine intermediates derived from 27 and 28 are also described in this paper.
0022-2623/79/ 1822-0850$01.00/0 0 1979 American Chemical Society
11 -Thiohomofolic Acid
Journal o f Medicinal Chemistry, 1979, Vol. 22, No. 7 851
Chart I
R
..
n
1; R I - O H ; XZ-NH
4;
2 ; R = - O H ;X = S
R R,-CH2-8-CH2-CH2 5;
lO;X=-CI 13; X=-N
14
12;X= -O-$-CH3 0
3
-S
R
- 0 - e -OCH3
R = -NOH R,=-NH2
7; R =
300 "C; yield -30% based on 35. The compound showed A, at 368 and 256 in 0.1 N NaOH with a shoulder at 276 nm. The UV absorption spectrum of 38 in 0.1 N HCl showed A, a t 298 and 250 nm and at pH 7.0 350 and 247 nm. It is noteworthy that the spectral characteristics of 11thiohomopteroic acid are drastically different from those of 10-thiopteroic acid in 0.1 N HCI. This compound showed relevant NMR signals in TFA at 8.78 (s, 1, Cy),8.07, 7.4 (2 d, 2, 2, aromatic) and 3.55 (t, 4, ethylene bridge) ppm. Anal. (Cl5HI3N5O3S)C, H, N, S. Alternate Synthesis of 11-ThiohomopteroicAcid (38). A mixture of 2.4 g (1.01 mmol) of 2,5,6-triamino-4-pyrimidinol and 2.72 g (2 mmol) of sodium acetate in 500 mL of DMF was deaerated by bubbling through N2 for 15 min. A solution of 2.54 g (10 mmol) of the hydroxy ketone 11 in 100 mL of DMF was also deaerated by this procedure. The two solutions were mixed, and stirring was continued under an atmosphere of N2 for 4 h. After this period, nitrogen was cut off and the stirring was carried out under aerobic conditions for 18 h. Then the mixture was refluxed for 2 h and DMF was removed under vacuum. The dark-brown residue thus obtained was triturated with water and filtered. The precipitate thus obtained after washing several times with distilled water was hydrolyzed as described previously for 37. The yield of the crude pteroic acid fraction was 400 mg. The mixture was then chromatographed on DEAE-cellulose. All fractions containing 38 were pooled, concentrated to a small volume, and acidified, and the precipitate thus obtained was rechromatographed twice to obtain pure 38,yield 60 mg. The U\:and NMR spectra of this compound was identical with those of 38 prepared by the unambiguous procedure in all respects. The UV and NMR spectra of this compound was identical with those of 38 prepared by the unambiguous procedure in all respects. In addition, alkaline permanganate oxidation of this material, according to a literature procedure previously employedlg for the oxidation of 6-alkylpteridines, gave pteridine-6-carboxylic acid, which was identical in all respects with an authentic sample prepared by the oxidation of folic acid. Preparation of 11-Thiohomofolic Acid (2) from 11Thiohomopteroic Acid (38). Method A. Solid-Phase Coupling Procedure. A solution of 171.5 mg (0.5 mmol) of 38 was made in 25 mL of dry MezSO by heating to 100 "C. This solution was cooled to room temperature and an equal volume of T H F was added. The mixture was chilled in an ice bath and treated with 0.07 mL (0.625 mmol) of freshly distilled hr-methylmorpholine. After keeping 15 min at this temperature, 0.065 mL (0.5 mmol) of freshly distilled isobutyl chloroformate was added. After 15 min at this temperature, the mixed anhydride was allowed to couple with an excess of resin-bound L-glutamate for 18 h as described previously from this laborator Cleavage of the product from the resin by earlier procedures using 2 N NaOH and dioxane and subsequent chromatography gave only 17 mg of the desired product. Most of the pteridine
Nair e t al. fractions were decomposition products. Modifications of this cleavage procedure, substituting acetone for dioxane, did not improve the yield of the product. Subsequent experimentations revealed that neither the pteroate nor the homofolate analogue is stable toward 2 N NaOH under the cleavage conditions, and therefore this method of coupling was abandoned. Method B. The mixed anhydride 39 from 0.5 mmol of 38 was made in Me,SO at room temperature without the use of THF. To this, a solution of 1 mmol of diethyl L-glutamate hydrochloride dissolved in IO mL of MezSO and 0.113 mL (1 mmol) of Nmethylmorpholine was added and stirred for 18 h. After this period, the reaction mixture was made 0.33 N with respect to NaOH by the addition of 1 N NaOH and hydrolyzed for 4 h. The pH of the hydrolysate was adjusted to 7.2 and the mixture chromatographed on a DEAE-cellulose column. Three products were eluted from the column. The least polar fraction was the monoethyl ester of 2 (NMR) and the most polar fraction was the desired product. The middle fraction consisted of unreacted starting material 38. The monoethyl ester could be rehydrolyzed to 2 in a solution of 0.33 N NaOH in acetonitrile for 4 h. The total yield of 2 obtained by this procedure was -40%. However, the recovered starting material could be recycled to get additional amounts of 2. The final product was isolated as a light yellow powder by the following procedure. The combined column effluents were concentrated to a small volume and acidified with glacial acetic acid, whereupon 2 precipitated from the solution. It was filtered and washed with distilled water, containing a trace of acetic acid, several times and dried under vacuum overnight at room temin 0.1 N NaOH perature over P,06. Compound 2 showed ,A, at 366 (e 7888)and 256 (t 29389) with a shoulder at 275 (E 19288) nm. In 0.1 N HCl it showed A, at 290 (t 14911)and 255 (c 15 151) nm and it showed relevant NMR signals at 8.80 (s, 1,C?),7.8, 7.45 (d, J = 9, 2, 2, aromatic), 3.55 (br, 4, ethylene), 2.85 and 2.5 (c, 4, glutamic acid) ppm, in complete agreement with the required structure. Anal. (CzoH2,NsO~S.5H~0) c , H,N, 0 , s. Preparation of 7,8-Dihydro-ll-thiohomofolic Acid. In a test tube, 47.2 mg of 2 (0.1 mmol) was suspended in 5 mL of distilled water and stirred. Very small amounts of KHCOBwere added to this mixture portionwise until a clear solution was obtained. A UV spectrum of this solution was recorded, which at 365; then this solution was treated with 50 mg showed a A, of sodium dithionite. After 5 min, the spectrum was recorded again and no appreciable change had occurred to the 365-nm peak. This treatment and observation of the spectrum was repeated until the A,, at 365 nm completely disappeared. At this point, which required 200 mg of dithionite and a duration of 20 min at room temperature, the solution was cooled in an ice bath and a few drops of glacial HOAc were added until the solution became acidic (pH -4.5). The precipitated dihydro derivative was filtered, washed three times with water, and dried under vacuum: yield 35 mg: U V A,, (0.1 N NaOH) 327, 285 nm. The ratio of the optical density of the 285-nm peak to the 327-nm peak was 2.71. When the reduction was carried out above room temperature, 2 was converted to the tetrahydro derivative identified by comparison with an authentic sample prepared by catalytic reduction (vide infra) of 2. Methods Used for Biological Testing. The preparation of dl,J,-tetrahydro-11-thiohomofolicacid was carried out as described previously for the preparation of dl,L-tetrahydrofolic acid using platinum oxide and hydrogen at atmospheric pressure. The catalytic reduction product showed a single UV absorption maximum at 290 nm when run at pH 7.4. The material was homogeneous by DEAE-cellulosechromatography. The dithionite reduction product obtained by treating 2 with dithionite above 40 "C was identical with this catalytic reduction product. When 7,8-dihydro-ll-thiohomofolate was substituted for the natural substrate in the assay medium previously described for the assay of L. casei DHFR, it was reduced by NADPH to the tetrahydro derivative. The velocity of this reduction was approximately 25% of 7,8-dihydrofolic acid. Dihydrofolate reductase,20thymidylate synthetase," and microbiological assays were carried out as de~cribed.'~
Acknowledgment. T h i s investigation was s u p p o r t e d by G r a n t s CH-53A from t h e American Cancer Society,
2-Acetylpyridine Thiosemicarbazones CA-10914 from the National Institutes of Health, and 780014 from the Alabama affiliate of the American Heart Association.
References and Notes (1) J. I. DeGraw, J. P. Marsh, Jr., E. M. Acton, 0. P. Crews, C. W. Mosher, A. N. Fujiwara, and L. Goodman, J. Org.
Chem., 30, 3404 (1965). (2) L. Goodman, J. I. DeGraw, R. L. Kisliuk, M. Friedkin, E. J. Pastore, E. J. Crawford, L. T. Plante, A. Nahas, J. F.
Morningstar, Jr., G. Kowk, L. Wilson, E. F. Donovan, and J. Ratzan, J. Am. Chem. SOC., 86, 308 (1964). (3) R. L. Kisliuk and Y. Gaumont, Chem. Biol. Pteridines,Proc. Znt. Symp., 4th, 1969, 357 (1970). (4) P. C. Crusberg, R. Leary, and R. L. Kisliuk, J. Biol. Chem., 245, 5292 (1970). ( 5 ) L. C. Mishra, A. S. Parmer, and J. A. R. Mead, Proc. Am. Assoc. Cancer Res., 11, 57 (1970). (6) J. A. R. Mead, A. Goldin, R. L. Kisliuk, M. Friedkin, L. Plante, E. J. Crawford, and G. Kowk, Cancer Res., 26,2374 (1966). (7) M. G. Nair and P. T. Campbell, J. Med. Chem., 19, 825 (1976). (8) M. G. Nair, P. C. O'Neal, C. M. Baugh, R. L. Kisliuk, Y. Gaumont, and M. Rodman, J. Med. Chem., 21,673 (1978). (9) H. R. Hornbeak and M. G. Nair, Mol. Pharmacol., 14, 299 (1978). (10) J. I. DeGraw, R. L. Kisliuk, C. M. Baugh, and M. G. Nair, J. Med. Chem., 17, 522 (1974).
Journal of Medicinal Chemistry, 1979, Vol. 22, No. 7 855 (11) M. G. Nair, P. T. Campbell, and C. M. Baugh, J. Org. Chem., 40, 1745 (1975). (12) M. G. Nair, P. T. Campbell, E. Braverman, and C. M. Baugh,
Tetrahedron Lett., 31, 2745 (1975). (13) G. F. Hennion and F. P. Kupiecki,J. Org. Chem., 18,1601 (1953). (14) W. E. Bachmann and W. S. Strive, Org. React., 1,38 (1942). (15) S. Y. Chen and M. G. Nair, J. Org. Chem., 43,4143 (1978). (16) Y. H. Kim, Y. Gaumont, R. L. Kisliuk, and H. G. Mautner, J. Med. Chem., 18, 776 (1975). (17) C. M. Baugh and E. Shaw, J. Org. Chem., 29,3610 (1964). (18) E. I. Fairburn, B. J. Magerlein, L. Stubberfield,E. Stapert, and D. I. Weisblat, J. Am. Chem. SOC.,76, 676 (1954). (19) E. L. R. Stokstad, B. L. Hutchings, J. H. Mowat, J. H. Boothe, C. W. Waller, R. B. Angier, J. Semb, and Y. Stubbarow, J. Am. Chem. SOC., 70, 7 (1948). (20) M. Chaykovsky, A. Rosowsky, N. Papathanosopoulos,K. N. Chen, E. J. Modest, R. L. Kisliuk, and Y. Gaumount, J. Med. Chem., 17, 1212 (1974). (21) A. J. Wahba and M. Friedkin, J. Biol. Chem., 237, 3794 (1962). (22) R. L. Blakley, Biochem. J.,65, 331 (1957). (23) R. L. Kisliuk, D. Strumpf, Y. Gaumont, R. P. Leary, and L. Plante, J. Med. Chem., 20, 1531 (1977). (24) A molecular ion having an m l e value of 404 had been inadvertently reported for this compound in the previous paper.15
2-Acetylpyridine Thiosemicarbazones. 1. A New Class of Potential Antimalarial Agents Daniel L. Klayman,* Joseph F. Bartosevich, T. Scott Griffin, Carl J. Mason, and John P. Scovill Walter Reed Army Institute of Research, Division of Experimental Therapeutics, Washington, D.C. 20012. Received January 8, 1979 Based on the antimalarial properties observed for 2-acetylpyridine4-phenyl-3-thiosemicarbazone (l),an extensive series of related thiosemicarbazones was prepared and tested against Plasmodium berghei in mice. Screening results indicated that the presence of the 2-pyridylethylidene group was critical and that certain phenyl, benzyl, phenethyl, or cycloalkyl groups at N4 of the thiosemicarbazone moiety also contribute to antimalarial activity. Thiosemicarbazones, a class of compounds possessing a wide spectrum of medicinal properties, have been studied for activity against tuberculosis,2l e p r ~ s ybacterial4 ,~ and viral5 infections, psoriasis! r h e ~ m a t i s mtrypanosomiasis,8 ,~ and coccidiosi~.~ In the past few years, thiosemicarbazones derived from 2-formylpyridine and related aldehydes have been of great interest because of their reported antineoplastic action.1° Among the thousands of compounds submitted for antimalarial screening by numerous contributors to the Division of Experimental Therapeutics have been several hundred thiosemicarbazides and thiosemicarbazones. Virtually all were devoid of activity, including the wellknown tuberculostat, p-acetamidobenzaldehyde 3-thiosemicarbazone (Thiacetazone, Tibione). One thiosemicarbazone, however, namely, 2-acetylpyridine 4-phenyl3-thiosemicarbazone (I),I1 attracted our attention because
QYH3
C=
i
NNHCNHCcH 5 1
it showed activity in our primary screen. It was decided
to exploit this interesting lead by ascertaining the molecular features essential for activity and utilizing them to develop a new class of antimalarial agents. The influence on biological action was observed when the structure of 1 was modified as follows: (1) the thiocarbonyl group was replaced by a carbonyl group; (2) the pyridine moiety was replaced by another heterocyclic, aromatic, or cycloaliphatic ring system; (3) the point of attachment of the ethylidene group to the pyridine ring was changed to the 3 and 4 positions; (4) the methyl of the ethylidene group was replaced by other alkyls or hydrogen; ( 5 ) the phenyl ring a t the terminal (N4) position of the thiosemicarbazone was replaced by various substituted phenyls, other cyclic structures, and various so-called antimalarial aliphatic side chains. This paper is one of the first to report on thiosemicarbazones possessing antimalarial activity.12 In it, we limit our discussion to those compounds which are monosubstituted a t N4 of the thiosemicarbazone moiety. Additional reports are in preparation which are devoted to related 2-acetylpyridine thiosemicarbazones that are disubstituted at N4and also to the antibacterial properties of this general class of compounds.
This article not subject to U.S. Copyright. Published 1979 by the American Chemical Society
